|
dCS Product Overview
| |||
 |
 |
 |
 |
|
|
||||||||
Chapter 5
Introducing CS-Engage
Figure 34: CS-Engage provides marketing assets.
CS-Engage helps marketers create intelligent, personalized shopping experiences for e-commerce shoppers. Use CS-Engage to:
- Gather and evaluate information about your site's visitors and customers.
- Divide your customers into demographic segments.
- Target those segments with personalized promotional offerings, product placements, or content recommendations.
Additional Asset Types for CS-Engage
CS-Engage works with basic assets or flex assets. CS-Engage provides an additional set of asset types, which are shown in the following table. As with other dCS products, you manage assets on your management site and then publish to your delivery site.
Table 6: Asset Types for CS-Engage
Developers and administrators create and manage the visitor data and recommendation assets, while marketers create and manage the segment and promotion assets. As with any of the other Content Server applications, you create and work with assets on the management site. Then, when assets are approved, you publish them to your delivery site.
Marketers and developers are expected to collaborate extensively to implement effective merchandising efforts.
Collect Visitor Data
Visitor data defines the kinds of information that you want to gather about visitors to your web site. You define the following different kinds of attributes to hold visitor data:
- Each visitor attribute holds one characteristic of a user. Examples of visitor attributes include
years_of_experience,job_category, andnumber_of_children. You might gather values for these attributes when a visitor registers on your web site.- Each history attribute holds one characteristic of something (perhaps a product).
History definitions are a set of history attributes. For example, consider a history definition named
purchasescomprising the history attributesSKU,itemname,quantity, andprice. CS-Engage treats each history definition as a single record and assigns a timestamp to each history definition.Use Segments to Categorize Visitors
Segments are assets that categorize groups of visitors based on the visitor data that you are gathering about them. You build segments by determining which kinds of visitor data to use as filtering criteria and then setting the values that qualify or disqualify a visitor for the segment. Marketers use the visitor data assets to create segments that define groups of visitors with one or more characteristic in common: geographic location, gender, job description, item purchased, and so on. You can define segments that are extremely broad (all first-time visitors, for example) or very focused (all first-time visitors who own RVs and live in Alaska).
You use the Segment Filtering forms in CS-Engage to categorize groups of visitors based on the visitor attributes, history attributes, and history definitions created by the developers.
Segments are the key to personalization with CS-Engage. When visitors browse your site, the information they submit is used to qualify them for segment membership. When the site displays a page with a recommendation or promotion, CS-Engage determines which segments a visitor belongs to and displays the product recommendations or promotional messages designated for those segments.
Recommend Products and Content
Recommendations are assets that determine which products or content should be featured or "recommended" on a site page. These assets are rules based on the segments the visitors qualify for, and, in some cases, relationships between the product or content assets.
Recommendations have templates. A recommendation returns a list of product or content assets to its template when the template is rendered on a site page.
There are three kinds of recommendations:
- Related items -- returns a list of recommended items based on relationships between flex assets that represent products and content. You determine which of these assets have these relationships.
- Static lists -- returns separate static lists of recommended items for selected segments, and with no segment affiliation.
- Dynamic lists -- returns a dynamic list of recommended items, based on an asset that you select.
Templates for Recommendations
Recommendations have templates. A recommendation returns a list of assets to its template when the template is rendered on a site page. The items in a list of recommended assets are rated according to their importance to the current visitor based on the segments that the visitor belongs to.
A recommendation collects, assesses, sorts, ranks, and then returns content or product assets for display. The template requests a certain number of assets. The recommendation determines which items meet its criteria and then returns those assets to the template. The template then formats and displays them in a specific location on the site page.
Two kinds of merchandising assets are displayed (rendered) on your online pages: recommendations and promotions. However, you use the same templates for both because promotions override or replace recommendations when they are running. If a promotion applies to the current visitor and the promotion was configured to override the recommendation, the template displays the promotion rather than the recommended assets it would normally display.
Run Promotions
Promotions are assets that define an offer of value (a discount) to the visitors based on the products that the visitor is buying and the segments that the visitor qualifies for. This value can be offered in several ways:
- A discount off the purchase price of the promoted products
- A discount off the entire value of the shopping cart
- A discount off shipping charges
- A combination discount: a shipping discount with a price or cart discount
During and after the promotion has run its course, you can analyze its effect on your visitors' buying habits. Then, you can decide whether to redeploy, refine, or discard it.
Templates for Promotions
Promotions do not have their own templates. You place a promotion on a site page by selecting a recommendation for it to replace or override. CS-Engage then uses the recommendation's template to render the promotion on the site page.
Analyze Data
Analysis Connector is a data collection application that supports CS-Engage. It enables you to capture web site events such as visitor activity and clickstream data, providing an HTTP-based interface for exporting that data from the Content Server database to a third-party analysis tool. You can then analyze and use that information to create segments for CS-Engage or make changes to your site based on the information that you gathered.
Analysis Connector collects data in a highly efficient manner. The event data that you want to capture normally occurs during high-traffic periods, which is not a time when you want to degrade performance by constantly writing to the database. Instead, Analysis Connector rapidly streams large volumes of data to a queue. As the queue fills up, the data is written to disk files. You configure a background process to load data from the disk files into Content Server database tables during off-peak hours.
Following are examples of events you might want to capture and analyze:
- Clicks and page views
- Visitor demographic information
- Product structure information
- Cart item information
- Abandoned cart information
- Order and order item information
Summary
You use CS-Engage to collect visitor data, segment your visitors into groups, recommend products and content, and run promotions. Segments are the key to personalizing your site pages. Your site gathers the visitor data that CS-Engage needs to determine which segments a visitor qualifies for. The visitors see merchandising and promotional messages based on their membership in those segments.
Two distinct functional groups--developers and marketers--must work closely together to implement effective merchandising efforts with CS-Engage. For information about the implementation of CS-Engage, see
Chapter 9, "CS-Engage: Sample Implementation."
|
|
||||||||||||||
